- From: Sandro Hawke <sandro@w3.org>
- Date: Thu, 04 May 2006 14:01:02 -0400
- To: "Peter F. Patel-Schneider" <pfps@inf.unibz.it>
- Cc: bry@ifi.lmu.de, public-rif-wg@w3.org
> > What does mean "Semantics: none"?
>
> Something like:
>
> The RIF will have no model theory. The RIF will have no proof theory. In
> sum, the RIF will just be concerned with rules as data - not as syntactic
> constructs in a logic.
>
> Compliance of a tool with the RIF will be defined as
> 1/ being able to accept (all) valid RIF documents;
> 2/ being able to emit (a reasonable number of) valid RIF documents;
> and
> 3/ accepting an emitted RIF document results in the same internal "state"
> that caused the emission.
> The last condition would be measured by identifying a number of RIF
> documents that are faithfully processed (i.e., no information is lost) and
> demonstrating that the emission of a RIF document after accepting one of
> these RIF documents into a clean system results in the document itself.
That design doesn't provide for interchange between rule systems,
though, does it?
I'm starting to think that the question "does RIF have semantics" is not
well formed. Perhaps each particular RIF dialect will have a formal
syntax and a formal semantics, and RIF on the whole will just be some
XML packaging machinery (corresponding to the RIFRAF ontology).
Hopefully there can be many modules shared between dialects, where both
the syntax and semantics are shared. I'm not sure if it'll ever make
sense to share syntax but not semantics for some part of a language.
-- Sandro
Received on Thursday, 4 May 2006 18:01:20 UTC